Docs / Syntax / Minimalist / Workspace / Coproduct / TraceGrading

Edge-count grading of the Δ^c bialgebra

import Linglib.Syntax.Minimalist.Workspace.Coproduct.TraceGrading · namespace ConnesKreimer

The graded subspaces of ConnesKreimer R (UnorderedTree (α ⊕ β)) under the edge-count grading (the summed UnorderedTree.numEdges of a forest), with the coproduct half of the grading compatibility (comulCAlgHomN_of'_mem_gradedSpan) — the graded content of Lemma 1.2.10.

TODO: once mathlib's graded coalgebra/bialgebra API lands (leanprover-community/mathlib4#39849), restate gradedPiece as a DirectSum.Decomposition and this file's content as a GradedBialgebra instance (with connectedness feeding the graded Hopf upgrade).

References

Edge-count grading

Per Marcolli et al. 2025 p. 37, Lemma 1.2.10:

Let V^c(𝔉_{SO_0}) denote the vector space (over ℚ) spanned by the workspaces F ∈ 𝔉_{SO_0}, endowed with the product given by the disjoint union ⊔ and the coproduct Δ^c of (1.2.8). The space V(𝔉_{SO_0}) is graded by the number of edges. Then (V^c(𝔉_{SO_0}), ⊔, Δ^c) is a graded bialgebra.

This file defines the graded subspaces and proves the coproduct half of the grading compatibility (comulCAlgHomN_of'_mem_gradedSpan); the product half is edge-count additivity over disjoint union (Multiset.map_add, Multiset.sum_add), and edge conservation through the trace cuts is cutSummandsCN_numEdges (Coproduct/TraceConservation.lean).

Homogeneous tensor span at fixed total edge degree

Δ^c preserves the edge-count grading (Marcolli et al. 2025 Lemma 1.2.10, p. 37).

theorem comulCAlgHomN_of'_mem_gradedSpanL139–159
{R : Type u_1} [CommRing R] {α : Type u_2} {β : Type u_3} (τ : UnorderedTree (α ⊕ β) → β) (F : UnorderedTree.Forest (UnorderedTree (α ⊕ β))) :
(comulCAlgHomN τ) (of' F) ∈ Submodule.span R {y : TensorProduct R (ConnesKreimer R (UnorderedTree (α ⊕ β))) (ConnesKreimer R (UnorderedTree (α ⊕ β))) | ∃ (i : ℕ) (j : ℕ) (_ : i + j = (Multiset.map UnorderedTree.numEdges F).sum) (xi : ConnesKreimer R (UnorderedTree (α ⊕ β))) (yi : ConnesKreimer R (UnorderedTree (α ⊕ β))), xi ∈ gradedPiece (α ⊕ β) i ∧ yi ∈ gradedPiece (α ⊕ β) j ∧ y = xi ⊗ₜ[R] yi}

Built from